prevention of failures I . . .

This image has an empty alt attribute; its file name is ed_02_fig_01_300x200.jpg




figure 1: the 5 cu m septic tank.

Continuing on in our series of articles with the theme “prevention of failures during service of structural plastic products” and in particular, underground tanks and chambers, in this article we appraise some of the common test methods applied to these products.

During the Nordic ARM conference held in Horsens, Denmark in February 2011, Martin and David, directors of Rota Design Limited, presented a case study involving the failure of a 5 cubic metre underground rotationally moulded PE septic tank. The tank didn’t fail, it imploded within hours of installation and many more followed. We presented the analysis of 100 projects taken a random from our files, which had been undertaken during recent years, and that alarmed us so much, it inspired us to write this series of articles.

The percentage that interested us most was the 19% of tanks and chambers that were intended for long term service underground which had failed in a matter of days or weeks after installation.

These designs were from third party designers, moulders and toolmakers and they all had one thing in common; they showed a complete lack of appreciation and understanding of the multitude and magnitude of the forces likely to be imposed upon the chamber wall during its intended lifetime, both in normal and abnormal service conditions.

The case study outlined the design of an apparently robust septic tank, its successful testing to an International Standard and its installation and subsequent catastrophic failure within hours.

The images also show how Rota Design Limited is able to replicate the failure mode and how the loss could have been avoided if we had been involved from the outset.

The sequence of images on the following pages is self explanatory.

              figure 2                                         figure 3

figure 2 shows the 3D CAD model and figure 3 shows the septic tank undergoing testing during the “sink test” and displaying only minor deformation under load i.e. a pass!


          figure 4                                       figure 5

figure 4 shows the Rota Design FEA meshed model ready for analysis studies whilst figure 5 shows the results of the FEA studies with hydrostatic pressure only applied, which confirms the result of the “sink test” i.e. with only minor deformation under load visible.

              figure 6                                               figure 7

figure 6 shows an FEA prediction of the tank failure when subjected to the short term effects of soil in combination with hydrostatic pressure.

figure 7 shows a photograph taken of the failure mode of a tank!

During the presentation to the Nordic ARM, David raised response from the delegates when he stated “We wish to thank all you designers out there who got it so wrong and gave us so much interesting work”! It’s not that we here at Rota Design Limited are arrogant or that we wish to rub people’s noses in the proverbial dirt, it’s just that we are passionate and totally committed to the reduction of harm being done to our industry on a worldwide basis by these failures. We see it as prodding the industry to act in a more responsible manner and that education and enlightenment of our designers is necessary so that they appreciate fully the forces acting on these structures.

So what lessons are to be learnt here? Firstly, burn your copy of the ISO standard which implies that an underground tank or chamber is safe if you apply the sink or vacuum test. Being immersed in water or applying vacuum does not replicate the multitude of forces acting on a structure when it is buried in soil or gravel in the long term. Rota Design Limited feel strongly that these standards should be withdrawn forthwith.

The interaction between polymers and soils is complex enough.  Now combine the effects of hydrostatic forces from groundwater, thermal loads, internal hydraulic or pneumatic loads, vehicular or pedestrian loads, surcharges from other sources and you may begin to appreciate the complex load cases and boundary conditions which may apply. And, which one set of cases is the worst case? We may not yet be able to truly replicate the real world in finite elements on our computer systems, but we at Rota Design Limited strive to model and predict the long term performance of these underground tanks and chambers as closely as modern technology will permit. At least then you have a reliable risk assessment on which to base your decision to order the tooling!